Bone Marrow Transplantation (2004) 34, 477–483 & 2004 Nature Publishing Group All rights reserved 0268-3369/04 $30.00 www.nature.com/bmt

Conditioning regimens High-dose treosulfan in patients with relapsed or refractory high-grade lymphoma receiving tandem autologous blood stem cell transplantation

M Koenigsmann1, M Mohren1, K Jentsch-Ullrich1, A Franke1, E Becker2, M Heim2, M Freund3 and J Casper3

1Clinic of Hematology/Oncology, Germany; 2Institute for Transfusion Medicine and Immunohematology, Germany; and 3University Magdeburg and Clinic of Hematology/Oncology, University Rostock, Germany

Summary: cytostatic agents include cyclophosphamide, etoposide, melphalan, BCNU, cytarabine and platinum compounds. This phase I/II study evaluated high-dose treosulfan in The alkylating drug busulfan has been used for 30 years in patients with high-grade lymphoma. In all, 21 patients high-dose therapy. It is characterized by profound stem cell (median age 51, 25–60 years) with primary refractory toxicity and cytotoxic activity in various hematological disease (n ¼ 3) or early (n ¼ 11) or late (n ¼ 7) relapse malignancies.3 Major drawbacks associated with busulfan received DexaBEAM and one course etoposide for include the high incidence of veno-occlusive disease cytoreduction and PBPC mobilization. Subsequently, 16 (VOD),4,5 a high variance of bioavailability, drug interac- patients received 30 g/m2 treosulfan and 140 mg/m2 tions associated with enzymatic activation in the liver6,7 and melphalan, followed by autologous transplantation. Nine finally a decreased threshold of seizure activity.8 patients received a 2nd high-dose treatment (HDT) with Treosulfan is a prodrug of a bifunctional alkylating 30 g/m2 treosulfan and 750 mg/m2 thiotepa. Recovery time cytotoxic agent with close structural similarity to busulfan,9 to 41/nl leukocytes and 425/nl thrombocytes was 8.9 which has been used for the treatment of ovarian cancer for (range 8–11) and 11.9 (8–16) days after 1st and 9.6 (7–13) many years.10,11 Myeloablative dose escalation upto 56 g/ and 13 (9–19) days after 2nd HDT. Reversible grade 3 or 4 m2 has recently been tested, and the maximum-tolerated nonhematologic toxicities included mucositis (n ¼ 7), dose (MTD) is 47 g/m2.12 Treosulfan can be given infection (n ¼ 7) and one episode of re-entry tachycardia. intravenously (i.v.) and is activated nonenzymatically, Two treatment-related deaths occurred after 2nd HDT. yielding highly reproducible pharmacokinetic profiles.13 Since three dose-limiting toxicities occurred among nine Owing to its antineoplastic activity and limited organ patients receiving tandem HDT, 30 g/m2 of treosulfan was toxicity, high-dose treosulfan has been considered for considered MTD in this setting. Patients with late autologous peripheral blood stem cell transplantation. compared to early relapse or refractory disease had a Recent in vitro data show activity in myeloma cells.14 higher probability of CR (6/7 vs 3/14 patients, P ¼ 0.017) Profound stem cell toxicity has been demonstrated in a and overall survival (71 vs 21%, Po0.05, 24–49 months preclinical model and it has been hypothesized that follow-up). In conclusion, high-dose treosulfan as major treosulfan, similar to busulfan, may also kill immature therapy component induces sustained complete remissions malignant progenitor cells.15 Most recently, treosulfan has in relapsed high-grade lymphoma with acceptable toxicity. been combined with fludarabine as a conditioning treat- Bone Marrow Transplantation (2004) 34, 477–483. ment in allogeneic stem cell transplantation.16 In the doi:10.1038/sj.bmt.1704626 following phase I/II trial for patients with relapsed and Published online 2 August 2004 refractory high-grade lymphoma, we tested the feasibility, Keywords: treosulfan; relapsed high-grade lymphoma; the MTD, and the efficacy of high-dose treosulfan in a autologous PBPC transplantation; tandem transplantation tandem high-dose approach.

High-dose chemotherapy with autologous stem cell trans- Materials and methods plantation can achieve long-term remission in patients with 1,2 relapsed high-grade lymphoma. Frequently employed Patient selection The study was open to adults up to 60 years of age with relapsed or refractory high-grade lymphoma who had Correspondence: Dr M Koenigsmann, Klinik fu¨ rHa¨ matologie und received chemotherapy with or without radiotherapy for Onkologie, Otto-von-Guericke-Universita¨ t, Leipziger Strasse 44, their primary lymphoma manifestation. Patients were Magdeburg 39120, Germany; E-mail: [email protected] considered refractory to the primary treatment if they Received 9 February 2004; accepted 9 May 2004 never achieved at least PR. Early or late relapse was defined Published online 2 August 2004 as relapse within or beyond 1 year after primary diagnosis.2 High-dose treosulfan in relapsed high-grade lymphoma M Koenigsmann et al 478 Patients with antecedent or concurrent neoplastic disease stem cell transplantation followed on day 0, and filgrastim were excluded. The protocol was started in October 1999 was given as described above. after approval by the local ethical committee. Patients were treated after giving written informed consent. Study design and toxicity evaluation

Induction therapy and stem cell mobilization One principle goal of the study was to evaluate dose- limiting toxicities (DLT) and to determine the MTD of For cytoreduction and stem cell mobilization (back-up), all high-dose treosulfan in combination with either melphalan patients were scheduled to receive one cycle of DexaBEAM or thiotepa. During hospitalization, clinical and laboratory with dexamethasone 3 Â 8 mg on treatment days 1–10, side effects were monitored daily and analyzed according to BCNU 60 mg/m2 on day 2, melphalan 20 mg/m2 on day 3, the CTC criteria of the National Cancer Institute (version etoposide 75 mg/m2 and cytarabine 100 mg/m2 every 12 h, 3.0). DLT was defined as any of the following toxicities: each on days 4–7. Filgrastim at 10 mg/kg/day was started on nephrotoxicity Xgrade 2, neurotoxicity Xgrade 2, other day 11 and was continued until the end of stem cell nonhematological toxicity Xgrade 3 (excluding alopecia, collection. Subsequently, another induction course was to nausea/vomiting, infections, mucositis grade 4 shorter than be given with etoposide at 2 g/m2 followed again by 7 days, and diarrhea grade III). The MTD was defined as filgrastim at 10 mg/kg/day for stem cell collection. Owing one dose level below the level at which greater than or equal to the additional in vivo purging effect, the second graft was to two of three or three of six patients experienced DLT, preferentially to be used for transplantation. At least respectively. 4 Â 106 CD34 þ cells/kg body weight (b.w.) had to be collected to provide for two stem cell transplantations. Response evaluation

Stem cell collection The other principle goal of the study was to evaluate the efficacy of the treatment at the MTD. Complete restaging As soon as the CD34 þ cell count exceeded 10/ml, PBPCs was performed including clinical and laboratory workup as were collected using a blood cell separator (Spectra, COBE well as a CT scan after the completion of the two sequential BCT Inc., Lakewood, CO, USA) with a semiautomated induction courses, immediately before the 2nd high-dose mononuclear cell collection program. The separation was cycle and 2 months after the last HDT. Complete remission continued until the necessary amount of CD34 þ cells was was defined as disappearance of all lymphoma manifesta- collected or until the number of circulating CD34 þ cells tions for at least 8 weeks. Partial response was defined as a fell below the 10/ml threshold. If possible, peripheral veins more than 50% reduction of all evaluable tumor lesions for were used; otherwise, a central venous catheter was placed a duration of at least 8 weeks. Time to progression and in the internal jugular vein. Blood flow was between 50 and survival time were calculated from the date of entry into the 90 ml/min, and the separation volume was the 2.5-fold of study protocol. the blood volume. The ration of whole blood to antic- Follow-up visits were scheduled for every 3 months oagulation ratio was kept regulated at 18:1, using mixed during the first year, every 4 months during the second anticoagulation with 3000 IU heparin to 600 ml ACDA year, and every 6 months thereafter. In addition to clinical (Fresenius, Hemocare). Cells were mixed with DMSO at a and laboratory testing, CT scans were performed twice a final concentration of 10% in freezing bags (Consarctic), year in the first 2 years and once a year thereafter. subjected to controlled-rate freezing (BV-25, Consarctic) and stored in liquid nitrogen under controlled standard conditions. Statistics The influence of relapse situation (refractory and early High-dose treatment relapse vs late relapse) on the achievement of complete remission was tested with Fisher’s exact test. Survival The first high-dose therapy consisted of melphalan 140 mg/ curves were calculated according to the Kaplan–Meier 2 m , given on day À4 as an i.v. infusion over 30 min, and method. Comparison of survival curves was made by the 2 treosulfan 30 g/m , given on day À3 as an i.v. infusion over log-rank test. 24 h. After the transplantation of at least 2 Â 106 CD34 þ cells/kg b.w. on day 0, filgrastim was given s.c. at 5 mg/kg/ day from day 1 until neutrophil recovery 41.5/nl. The þ Results second high-dose therapy course was to be given 8–12 weeks after the first one. It contained treosulfan and Patient characteristics (Table 1) thiotepa. Treosulfan was split into daily doses of 10 g/m2, and was given on days À5, À4, and À3, each time as an i.v. In all, 21 patients with a median age of 51 (25–60) years infusion over 2 h, yielding a total dose of 30 g/m2/day. It were included in the study. Three patients were refractory was planned to escalate the total dose of treosulfan in both to the primary treatment, 11 patients had early and seven high-dose cycles to 39 and 48 g/m2. Thiotepa was split into patients had late relapse. Four patients had peripheral T- daily doses of 250 mg/m2, and was given on days À5, À4, cell lymphoma; all other patients had B-cell lymphoma. For and À3, each time as i.v. infusion over 60 min, yielding a primary treatment, 14 patients had received radiotherapy in total dose of 750 mg/m2/day. Autologous peripheral blood addition to chemotherapy. One patient had front-line

Bone Marrow Transplantation High-dose treosulfan in relapsed high-grade lymphoma M Koenigsmann et al 479 Table 1 Patients characteristics

ID Sex Age Initial diagnosis Prior therapy TTR Disease (years) (months) status Histology Stage Chemotherapy RT Gy RT site Best response

1 M 53 CB IIEA CHOP  6 30 Supra PD 6 REF 2 M 32 Peripheral T-cell IIIEA CHOP  6 30 Supra PR 10 ER 3 F 52 FL III IIIB CHOEP  5 30 Supra CR 64 LR 4 M 57 DLC-AP IIIEA MegaCHOEP; 3  auto-tpl CR 8 ER 5 M 51 Peripheral T-cell IIAE CHOP  6 40 Sinus CR 9 ER 6 M 60 CB IIA CHOP  6 36 Supra CR 42 LR 7 M 46 CB IV CHOP  6PR6ER 8 M 58 IB IVB CHOP  6 36 Infra PR 5 ER 9 F 43 CB IV CHOEP  6 36 Infra PR 10 ER 10 M 52 FL II–III IVA COP  9 27 Infra CR 98 LR 11 F 55 CB/LCAP IIB CIVEP  6 36 Infra CR 35 LR 12 M 33 HD IIB CHOEP  6PR7ER 13 F 54 Peripheral T cell IVA CHOP  6 36 Supra +infra PR 8 ER 14 F 38 DLC-AP IV CHOEP  6 36 Supra PR 19 LR 15 M 34 Burkitt like IISA B-NHL-Prot  6 30 Infra PR 10 ER 16 F 25 CB IA CHOEP  4CR72LR 17 M 47 CB IVA CHOEP  6NC4REF 18 M 60 MCL blastoid variant IVA CHOEP  6CR12ER 19 F 54 CB IIA B-NHL-Prot  6 36 Supra CR 9 ER 20 M 51 Burkitt IIA CHOEP  3 CNS CRr 20 LR 21 F 43 CB IVB CHOEP  3PD2REF Median 51

CB ¼ diffuse large B-cell lymphoma, centroblastic variant; IB ¼ diffuse large B-cell lymphoma, immunoblastic variant; FL ¼ follicular lymphoma; DLC- AP ¼ diffuse large B-cell lymphoma, anaplastic large-cell variant; HD ¼ Hodgkin’s disease; peripheral ¼ peripheral T-cell lymphoma, unspecified; MCL ¼ mantle cell lymphoma; TTR ¼ time to relapse (from 1st diagnosis); ER ¼ early relapse, o1 year after primary diagnosis; LR ¼ late relapse, 41 year after primary diagnosis; REF ¼ refractory disease.

therapy with three intensified CHOEP regimens and PBSC grade 4 leukopenia and thrombocytopenia. The mean time support. of leukocyte recovery to 41/nl was 8.9 days (range 8–11 Initial cytoreductive treatment with DexaBEAM was days). The mean time of platelet recovery to 425/nl was given to 21 patients followed by one to four PBPC 11.9 days (range 8–16 days). Grade 3 infections occurred in aphereses (mean 2.3) in 12 patients starting between days four of 16 (25%) patients. Mucositis grade 3 was seen in 18 and 24 (median 20) of the cycle. The median yield was three patients, grade 4 in one patient. No further grade 4 7.3 (range 1.7–21.9) Â 106 CD34 þ cells per kg b.w. In four toxicities were noted besides hematotoxicity and alopecia. patients, stem cell collection had been performed earlier Other grade 3 toxicities included vomiting and/or diarrhea during the 1st CR and was not repeated. In three patients, in four patients. One patient developed Paget–von- stem cell collection was postponed due to grade 3 infection Schroetter’s syndrome at the previous site of her central at the time of their CD34 þ cell peak levels. In these venous catheter after being discharged from the hospital. patients, apheresis was performed after etoposide treatment She subsequently received coumarin anticoagulation for 3 only. Two patients progressed under DexaBEAM and months. discontinued the study before PBPC apheresis. One patient developed secondary acute myeloid leukemia after PBPC 2nd High-dose therapy (Table 2b) apheresis and did not proceed to autologous transplantation. A second cytoreductive therapy cycle with etoposide was At 6 to 11 weeks (mean 8.3) after the 1st high-dose therapy, given to 15 patients. One patient received an alternative nine of 16 patients received a 2nd high-dose therapy. Of the MTX-containing polychemotherapy regimen instead of remaining seven patients, four had progressed after the 1st monotherapy with etoposide due to progression after HDT, and three patients refused a 2nd HDT. The patients DexaBEAM. In eight patients, two stem cell collections who continued the study received a median of 3.5 (range were performed between days 11 and 20 (median 15) of 1.7–32.6) Â 106 CD34 þ cells per kg b.w. Grade 4 etoposide treatment. The median yield was 4.6 (range leukopenia and thrombocytopenia was seen in all patients 0.8–32.6) Â 106 CD34 þ cells per kg. and hematological recovery occurred after a mean of 9.6 days (range 7–13 days) for 41/nl leukocytes and after a 1st High-dose therapy (Table 2a) mean of 13.0 days (range 9–17 days) for 425/nl platelets. Unfortunately, two patients died at days þ 17 and þ 19, A total of 16 patients received high-dose therapy with both with pneumonia and grade 4 sepsis. One of these treosulfan and melphalan. The autologous peripheral blood patients had previously undergone three high-dose regi- stem cell transplants contained a median of 4.4 (range mens during front-line therapy. In addition, the interval 1.7–21.9) Â 106 CD34 þ cells per kg b.w. All patients had between the beginning of the first and the second HDT was

Bone Marrow Transplantation High-dose treosulfan in relapsed high-grade lymphoma M Koenigsmann et al 480 Table 2 (a) 1st High-dose therapy, individual recovery and toxicity and (b) 2nd high-dose therapy, recovery and toxicity

ID Transplant quality Recovery time Toxicity (numbers represent individual CTC grade) (days)

CD34+ cells/kg WBC41PLT425 Hb WBC PLT Bleeding Infection Mucositis Other (only 1III)

(a) 1 4.8 9 13 2 4 4 0 1 1 Diarrhea 2 9 12 2 4 4 0 Herpes 1 3 2 8 8 1 4 4 0 2 2 Diarrhea 4 4.3 8 11 1 4 3 0 3 3 5 1.7 9 13 1 4 4 0 0 1 6 2.6 9 13 2 4 4 0 0 2 8 4.8 9 11 4 4 4 1 3 0 9 4.4 9 11 3 4 4 0 3 2 Vomiting 10 21.9 8 10 2 4 4 0 0 1 11 6 9 16 0 4 4 0 3 4 12 2.3 9 11 0 4 4 0 0 0 14 2.4 8 10 2 4 4 0 2 3 16 8.4 9 10 3 4 4 0 0 2 17 8.7 11 19 4 4 4 0 1 3 Vomiting, diarrhea 20 3.8 9 10 3 4 4 0 2 0 21 6.4 9 13 3 4 4 0 2 2 Median: 4.4 Mean 8.9 Mean 11.9

ID Weeks 1st Transplant Recovery time Toxicity (numbers represent individual CTC grade) to 2nd HDT (days)

CD34+ cells/kg WBC41PLT425 Hb WBC PLT Bleeding Infection Mucositis Other (13or4)

(b) 2 6 5.2 9 17 3 4 4 0 4 2 14 fatal pneumonia day +17 4 6 3.4 13 19 1 4 4 0 4 3 14 fungal pneumonia, fatal sepsis day +19 5 8 1.7 9 2 4 4 0 1 4 13 diarrhea, sv re-entry tachycardia 8 10 2.1 9 12 2 4 4 0 0 2 13 liver TA (1 day) 9 8 5.1 11 13 3 4 4 1 3 1 13 vomiting, diarrhea 10 8 32.6 7 9 0 4 4 0 0 0 12 10 2.1 10 11 2 4 4 0 2 1 14 11 4.5 10 14 4 4 4 1 3 4 20 8 3.5 9 13 2 4 4 0 2 2 Median: 3.5 Mean 9.6 Mean 13

only 6 weeks, which may have contributed to an increased course. Six patients entered CR after the 1st, and one susceptibility towards infectious complications. Another patient after the 2nd high-dose therapy course. Further three patients developed grade 3 neutropenic fever. One of analysis revealed that none of the three patients with these patients experienced a reversible grade 3 re-entry refractory disease, and only three of 11 patients with early tachycardia during septicemia on day þ 9, which was relapse ever reached CR. In contrast, six of seven patients pharmacologically controlled by an i.v. bolus of adenosine. with late relapse (85%) entered CR. Fisher’s exact test Another two patients suffered from grade 3 vomiting with showed a significant difference between the poor risk group or without diarrhea. One patient showed a reversible grade (refractory disease and early relapse) and the low-risk 3 ALAT elevation only on day –1. Throughout the study group (late relapse) with respect to the achievement of CR no VOD occurred. (P ¼ 0.017). Five of nine patients in CR (23.8% of the total Taking together the two treatment-related deaths and the study group) remain progression free after 25 to 48 months observed re-entry tachycardia during the 2nd HDT, there (median 35 months). were three DLT among the nine patients who received both high-dose regimens. It was therefore decided not to escalate Overall survival (Figure 1) the treosulfan dosage and to consider 30 g/m2 of treosulfan as the MTD in this tandem setting. After a median follow-uptime of upto 24 months (range 18–49 months), the Kaplan–Meier plot showed a prob- Response (Table 3) ability of survival of 46%. The curves of overall survival were significantly different for patients with late vs early Altogether nine of 21 patients reached CR (42.8%). In two relapse or refractory disease (P ¼ 0.0022). The probability of of these patients, CR was documented after the etoposide long-term survival was 71.4% after a follow-uptime of at

Bone Marrow Transplantation High-dose treosulfan in relapsed high-grade lymphoma M Koenigsmann et al 481 Table 3 Response to study treatment

ID Response to treatment Completion of protocol treatment Follow-up Relapse after study treatment

DexaBEAM Etoposid 1st 2nd Until cycle Reason to TTP Survival 3rd line treatment Reponse HDT HDT stop protocol (months) (months)

1 No PD No PD PD 1st HDT PD 2 12 2 No PD No PD PD 2nd HDT Death 3 4 3 No PD No PD CR 1st HDT Patients will 8 49+a CLAD CR 4 No PD No PD PD 2nd HDT Death 4 5NoPDb No PD CR 2nd HDT 48+ 6 No PD No PD PD 1st HDT PD 6 10 Dexamethasone, Taxotere PD 7 PD DexaBEAM PD 0 2 DHAP PD 8 No PD PR PR PD 2nd HDT 8 9 9 No PD PR PR CR 2nd HDT 36+ 10 No PD PR CR 2nd HDT (6)c 35+ Fludara/Bendamustin CR 11 No PD PR CR 1st HDT Patients will 33+ 12 No PD CR 2nd HDT 6 32+ Radiotherapy CR 13 PD DexaBEAM PD 2 4 14 No PD PR CR 2nd HDT 16 29+ Fludara/Benda/Ritux PD 15 PD DexaBEAM PD 0 1 DHAP PD 16 No PD CR 1st HDT Patients will 25+ Rituximab CR 17 No PD PD PD 1st HDT PD 5 8 3 Regimens PD 18 PD DexaBEAM PD 2 2 3 Regimens PD 19 No PD n.e.d DexaBEAM Secondary AML 2 20 No PD PD CR 2nd HDT 7 8 Fludara/Benda/Ritux PD 21 PD PR after A1 PR 1st HDT PD 4 4 Dexamethasone PD 2 CR 5 CR 1 CR 8 PD Median 8 aPatient alive at 49 months follow-upis indicated by 49+. bDifferent induction protocol (2 Â BMPD). cOngoing remission of the aggressive lymphoma component but relapse of the grade 1 lymphoma component. dNot evaluable due to prior splenectomy. least 24 months and was significantly higher (Po0.05) than The rationale to choose a tandem high-dose approach for patients with early relapse or refractory disease (21.4%). was derived from our intention to combine treosulfan with only one other cytotoxic agent at the same time. In order to avoid undertreatment, a second high-dose cycle was added. Discussion Although one patient entered CR only after the 2nd HDT, this strategy did not allow to conclude whether the response To the best of our knowledge, this is the first published duration in our patients was primarily due to the activity of clinical study using high-dose treosulfan and autologous a single high-dose regimen or to the entire tandem high- stem cell transplantation in a lymphoma patient popula- dose approach. tion. Within a phase I trial, testing high-dose treosulfan, The dosage of treosulfan of 30 g/m2 for each cycle was one CR and one PR have been previously described among extrapolated from a phase I trial where 47 g/m2 was the two non-Hodgkin’s lymphoma patients, in addition to one MTD for a single, stem cell supported, therapy with PR in a patient with Hodgkin’s lymphoma.12 treosulfan only. Since there was no data on tandem Treosulfan was the major component of this protocol transplants with treosulfan, it was decided to start with a since it was the only drug used in both high-dose regimens. total dosage of no more than 60 g/m2 for both therapy Nevertheless, it is hard to appreciate its exact role since it cycles. Given a total of three DLT in nine patients who had was combined with either melphalan or thiotepa. While received tandem therapy, this decision has proven to be melphalan alone is very potent in the treatment of reasonable. Owing to the sequential application of two myeloma, there is little data on its single-drug activity in high-dose regimens, however, this protocol did not allow high-grade lymphoma. A recently published comparative the evaluation of the MTD of treosulfan within a single study tested melphalan vs cyclophosphamide in a BCNU-, high-dose combination protocol. etoposide-, and cytarabine-based protocol. The melphalan- With respect to the bioequivalency of our doses of containing regimen seemed to be more toxic to the treosulfan in comparison to other alkylators in hematologic gastrointestinal tract; however, the antilymphoma activity malignancies, one study has compared the in vitro activity appeared to be similar for both combinations.17 High-dose of treosulfan and melphalan.14 At a concentration of thiotepa is used in combination protocols with other 100 mmol/l, both drugs induced cell death in approximately alkylating drugs like busulfan and cyclophosphamide for 60% of primary myeloma cells. Similar concentrations were patients with lymphoma or multiple myeloma.18,19 The only reached in the serum of patients treated with a melphalan way to assess the contribution of a single drug is to dosage of 200 mg/m2. However, the serum levels of patients compare different combination regimens for efficacy and who were treated with 30–47 g/m2 treosulfan were several toxicity. This is a general problem in treatment situations in fold higher.12 With all due caution, it can therefore be which combination protocols are used. assumed that the biologic activity of the treosulfan dosage

Bone Marrow Transplantation High-dose treosulfan in relapsed high-grade lymphoma M Koenigsmann et al 482 a therapy should be excluded from tandem transplantation. 1.0 We do think that tandem transplantation is to be further explored in this setting, and it should currently be restricted to those patients who reach only a PR after the first HDT. 0.8 Interestingly, one of our patients with early relapse reached CR only after the second high-dose therapy, and she is in continuous CR for meanwhile 33 months. 0.6 Another toxic problem was mucositis, which is a strain on the patient’s quality of life and is associated with a high risk of infections. While etoposide is one of the most

Probability 0.4 efficient drugs for stem cell mobilization21,22 and a component of combination protocols in high-grade lym- 23 0.2 phoma, its use is clearly associated with the occurrence of post-transplant mucositis.24 The observation of two cases of grade 4 mucositis in our series is consistent with this 0.0 finding and will be taken into account for future trials. The 0 12 24 36 48 60 nonhematologic DLTs observed during a phase I trial with Survival time [months] treosulfan monotherapy were diarrhea, mucositis, acidosis, 12 b reversible skin reactions, vomiting, and headache. It 1.0 remained unclear as to whether the observed re-entry tachycardia during neutropenic fever in one of our patients after the 2nd HDT was due to a toxic effect or due to a pre- 0.8 existing cardiac problem. Confirming the early data of Philipand of the PARMA trial,1,25 we saw a superior outcome of patients with late 0.6 relapse reaching a CR rate in six of seven cases (85.7%) and an ongoing median survival of 35 months (25–48 months) groups in five of seven patients (71.4%). Owing to the small

Probability 0.4 Late relapse number, however, this finding has to be looked at with caution. censored The main problems in our series arose in patients with 0.2 Refractory or primary refractory disease and early relapse. In keeping early relapse with the data of Guglielmi et al,25 only three of 11 patients

0.0 censored with early relapse entered CR. Disease progression 0 12 24 36 48 60 prevented four patients each from proceeding to the 1st Survival time [months] or to the 2nd high-dose therapy cycle. Two of our patients were chemorefractory at the time of transplantation, but Figure 1 Kaplan–Meier plot showing probability of overall survival were able to receive high-dose therapy. One of these (46%) for all patients entering the study (a), and separately for patients with late relapse (at least 1 year from 1st diagnosis) or early relapse and patients progressed further. However, the other patient refractory disease (b). The course of the curves is significantly different (log- reached complete remission after the 1st HDT, and he even rank test, P ¼ 0.0022). Probability of long-term remission (at least 24 received a 2nd HDT. These data are too limited to draw months) is significantly higher for patients with late relapse (71.4%) conclusions on the activity of our protocol for patients who compared to patients with early relapse or refractory disease (21.4%, are chemoresistent at relapse, but further study of this Po0.05). question is urgently needed. Interestingly, Vose et al26 reported data from the ABMTR (Autologous Blood and used in our study was not inferior to the activity of other Marrow Transplant Registry) where patients who never alkylators. achieved remission with conventional therapy had a 44% Lee et al20 showed a 28% incidence of VOD in patients CR rate, and a probability of progression-free and overall receiving a combination of busulfan, melphalan, and survival at 5 years after transplantation of 31%. thiotepa. At variance, we did not observe any VOD in A contribution for the treatment of refractory disease our patients treated with treosulfan. On the other hand, might be given by allogeneic transplantation due to the two fatal infectious complications in our study after the graft-versus lymphoma effect.27–29 Further, the monoclonal second high-dose cycle was a serious problem. In both antibody rituximab has the potential to improve the cases, the interval between the two high-dose treatments treatment results in relapsed or refractory B-cell lymphoma was short with only 6 weeks. One patient carried an on several levels; first, at initial cytoreductive treatment for additional toxic risk since he had previously received three rapid induction of remission, second, for in vivo stem cell dose-intensified regimens with autologous transplantation purging, and lastly, during or after HDT for the during primary therapy. These additional risk factors maintenance of remission.30–32 should be avoided in the future. To this purpose, the In conclusion, the feasibility to include treosulfan into a interval between the high-dose regimens should be extended high-dose combination regimen has been demonstrated. to at least 8 weeks, and patients with prior high-dose Further, we have defined the MTD of treosulfan in this

Bone Marrow Transplantation High-dose treosulfan in relapsed high-grade lymphoma M Koenigsmann et al 483 tandem setting. Our data show low organ toxicity and A single centre analysis on toxicity and efficacy. Leukemia promising activity in patients, especially those with late Lymphoma 2003; 44: 1151–1158. relapse of aggressive lymphoma. Therefore, further evalua- 18 Shimoni A, Smith TL, Aleman A et al. Thiotepa, busulfan, tion of treosulfan in patients with lymphoma and other cyclophosphamide (TBC) and autologous hematopoietic hematologic malignancies is warranted. transplantation: an intensive regimen for the treatment of multiple myeloma. Bone Marrow Transplant 2001; 27: 821–828. 19 Cheng T, Forsyth P, Chaudhry A et al. High-dose thiotepa, References busulfan, cyclophosphamide and ASCT without whole-brain radiotherapy for poor prognosis primary CNS lymphoma. 1 PhilipT, Guglielmi C, Hagenbeck A et al. Autologous bone Bone Marrow Transplant 2003; 31: 679–685. marrow transplantation as compared with salvage chemother- 20 Lee JL, Gooley T, Bensinger W et al. Veno-occlusive disease of apy in relapses of chemotherapy-sensitive non-Hodgkin’s the liver after busulfan, melphalan, and thiotepa conditioning lymphoma. N Engl J Med 1995; 333: 1540–1545. therapy: incidence, risk factors, and outcome. Biol Blood 2 Bosly A, Gisselbrecht C. Therapy intensification with auto- Marrow Transplant 1999; 5: 306–315. logous transplantation in non-Hodgkin’s lymphomas. Bull 21 Reiser M, Jostin A, Draube A et al. Successful peripheral Cancer 2001; 88: 877–887. blood stem cell mobilization with etoposide in patients 3 Gupta V, Lazarus HM, Keating A. Myeloablative condition- with relapsed or resistant lymphoma who failed cyclopho- ing regimens for AML allografts: 30 years later. Bone Marrow sphamide mobilization. Bone Marrow Transplant 1999; 23: Transplant 2003; 32: 969–978. 1223–1228. 4 Jones RJ, Lee KS, Beschorner WE et al. Venoocclusive disease 22 Junghanss C, Leitha¨ user M, Wilhem S et al. High-dose of the liver following bone marrow transplantation. Trans- etoposide phosphate and G-CSF mobilizes peripheral blood plantation 1987; 44: 778–783. stem cells in patients that previously failed to mobilize. Ann 5Ha¨ nel M, Kro¨ ger N, Sonnenberg S et al. Busulfan, cyclopho- Hematol 2001; 80: 96–102. sphamide, and etoposide as high-dose conditioning regimen in 23 Moskowitz CH, Bertino JR, Glassman JR et al. Ifosfamide, patients with malignant lymphoma. Ann Hematol 2002; 81: carboplatin, and etoposide: a highly effective cytoreduction 96–102. and peripheral-blood progenitor-cell mobilization regimen for 6 Slattery JT, Risler LJ. Therapeutic monitoring of busulfan in transplant-eligible patients with non-Hodgkin’s lymphoma. hematopoietic stem cell transplantation. Ther Drug Monit J Clin Oncol 1999; 17: 3776–3785. 1998; 20: 543–549. 24 Bolwell BJ, Kalaycio M, Sobecks R et al. A multivariable 7 Hassan M. The role of busulfan in bone marrow transplanta- analysis of factors influencing mucositis after autologous tion. Med Oncol 1999; 16: 166–176. progenitor cell transplantation. Bone Marrow Transplant 8 De La Camera R, Thomas JF, Figuera A et al. High dose 2002; 30: 587–591. busulfan and seizures. Bone Marrow Transplant 1991; 7: 25 Guglielmi C, Gomez F, PhilipT et al. Time to relapse has 363–364. prognostic value in patients with aggressive lymphoma 9 Feit PW. Stereoisomer 1,4-di-O-methansulfonyl-butan-1,2,3,4- enrolled in the PARMA trial. J Clin Oncol 1998; 16: tetrole. Tetrahedron Lett 1961; 20: 716–717. 3264–3269. 10 Fenelly J. Treosulfan (hydroxybusulfan) in ovarian carcinoma. 26 Vose JM, Zhang MJ, Rowlings PA et al. Autologous In: Nelson JD, Grassi C (eds). Current Chemotherapy and transplantation for diffuse aggressive non-Hodgkin’s lympho- Infectious Disease. American Society for Microbiology: ma in patients never achieving remission: a report from the Washington, DC, 1980; pp 1683–1684. Autologous Blood and Marrow Transplant Registry. J Clin 11 Meden H, WittkopY, Kuhn W. Maintenance chemotherapy Oncol 2001; 19: 406–413. with oral treosulfan following first-line treatment in patients 27 Seropian S, Bahceci E, Cooper DL. Allogeneic peripheral with advanced ovarian cancer: feasibility and toxicity. Antic- blood stem cell transplantation for high-risk non- ancer Res 1997; 17: 2221–2224. Hodgkin’s lymphoma. Bone Marrow Transplant 2003; 32: 12 Scheulen ME, Hilger RA, Oberhoff C et al. Clinical phase I 763–769. dose escalation and pharmacokinetic study of high-dose 28 Bierman PJ, Sweetenham JW, Loberiza FR et al. Syngeneic chemotherapy with treosulfan and autologous peripheral hematopoietic stem-cell transplantation for non-Hodgkin’s blood stem cell transplantation in patients with advanced lymphoma: a comparison with allogeneic and autologous malignancies. Clin Cancer Res 2000; 6: 4209–4216. transplantation – The Lymphoma Working Committee of the 13 Hilger RA, Harstrick A, Eberhardt W et al. Clinical International Bone Marrow Transplant Registry and the pharmacokinetics of intravenous treosulfan in patients with European Group for Blood and Marrow Transplantation. advanced solid tumors. Cancer Chemother Pharmacol 1998; 42: J Clin Oncol 2003; 21: 3744–3753. 99–104. 29 Mollee P, Lazarus HM, Lipton J. Why aren’t we performing 14 Meinhardt G, Dayyani F, Jahrsdorfer B et al. Treosulfan is an more allografts for aggressive non-Hodgkin’s lymphoma? Bone effective inducer of cell death in myeloma cell lines and Marrow Transplant 2003; 31: 953–960. primary myeloma cells from patients. Br J Haematol 2003; 122: 30 Gisselbrecht C, Mounier N. Rituximab: enhancing outcome of 892–899. autologous stem cell transplantation in non-Hodgkin’s lym- 15 Fichtner I, Becker M, Baumgart J. Antileukaemic activity of phoma. Semin Oncol 2003; 30: 28–33. treosulfan in xenografted human acute lymphoblastic leukae- 31 Benekli M, Hahn T, Shafi F et al. Effect of rituximab on mias (ALL). Eur J Cancer 2003; 39: 801–807. peripheral blood stem cell mobilization and engraftment 16 Casper J, Knauf W, Kiefer T et al. Treosulfan and fludarabine: kinetics in non-Hodgkin’s lymphoma patients. Bone Marrow a new toxicity-reduced conditioning regimen for allogeneic Transplant 2003; 32: 139–143. hematopoietic stem cell transplantation. Blood 2003; 103: 32 Shimoni A, Hardan I, Avigdor A et al. Rituximab reduces 725–731. relapse risk after allogeneic and autologous stem cell trans- 17 Jantunen E, Kuittinen T, Nousiainen T. BEAC or BEAM for plantation in patients with high-risk aggressive non-Hodgkin’s high-dose therapy in patients with non-Hodgkin’s lymphoma? lymphoma. Br J Haematol 2003; 122: 457–464.

Bone Marrow Transplantation